Vithursan Thangarasa's picture

6 2 11

Vithursan Thangarasa

vithursant

·

https://vithursant.com/

AI & ML interests

Large language models, Sparse Neural Network training, Generative Computer Vision

Recent Activity

new activity 3 days ago

cerebras/GLM-4.5-Air-REAP-82B-A12B:FP8 quant?

new activity 3 days ago

cerebras/GLM-4.5-Air-REAP-82B-A12B:A little confused on memory usage (vLLM newbie)

liked a model 5 days ago

bartowski/cerebras_GLM-4.5-Air-REAP-82B-A12B-GGUF

View all activity

Organizations

upvoted a collection 9 days ago

Cerebras REAP

Sparse MoE models compressed using REAP (Router-weighted Expert Activation Pruning) method • 11 items • Updated 2 days ago • 29

upvoted a paper 11 days ago

REAP the Experts: Why Pruning Prevails for One-Shot MoE compression

Paper • 2510.13999 • Published 14 days ago • 4